Comparison of digital camera formats

 Return to Home Page

Since I changed from film to digital photography, I have used several cameras. One major factor that distinguishes different digital cameras is the size of the sensor that is used to capture the image.

Back in 'film days', almost everyone used 35mm film and the question of format size rarely arose. Small formats, like the short-lived '110', gave poor results, while landscape photographers, seeking smooth gradation of tones, chose 'medium format' cameras. Whichever format was used, there was a selection of different types of film to choose from, depending on whether fine-grain or high-speed were the priority.

When digital photography arrived, everything changed and, for reasons of cost, the digital sensors were initially much smaller than a frame of 35mm film. Most cameras used sensors that were very much smaller even than a '110' frame, while 'up market' DSLRs were only 'APS'-size, or around 'half frame' in film terms.

Photographers began to notice that the images from digital cameras looked different from their previous film images and it became apparent that the reasons behind these differences were quite complex. Of course, there was no real reason why a digital sensor should be the same size as a film frame, since the mechanisms by which the image is generated are quite different between the two technologies. Nevertheless, those photographers who used interchangeable-lens cameras wanted to be able to continue to use the lenses with which they were familiar and so, the concept of 'equivalent' focal length emerged.

Angle of View

Initially, this 'equivalence' was based solely on the angle of view recorded by any given lens. For example, a 'wide angle' lens for a 35mm film camera might have a focal length of 28mm but, to obtain the same angle of view on APS format requires an 18mm focal length. At the other end of the scale, the angle of view of a 300mm lens on a film camera could be achieved with only a 200mm lens on an APS digital camera and even less on cameras with smaller sensors. This was seen as an advantage by wildlife photographers, who could get larger images from smaller and cheaper lenses.

When I changed from using Nikon (APS-format) cameras to Olympus (micro 4/3), I was able to take advantage of the smaller sensor to buy lighter and more compact lenses with the same telephoto 'reach'. Using a 300mm lens with a 1.4X teleconverter on my Nikon gives an 'equivalent' focal length of 630mm, whereas the zoom lens on my Olympus has an 'equivalent' range of 200 – 800mm, when compared with the field of view of a 35mm film camera.

Telephoto comparison

Olympus E-M1 with 100-400mm lens (weight 1.6 kg) compared with
Nikon D300s with 300mm lens +1.4X TC (weight 2.4 kg)


Depth of Field

For some applications, such as portraiture, 'depth of field' is important, because it enables the photographer to 'isolate' the subject or to draw attention to a particular part of the image. It follows that 'equivalence' means rather more than simply the angle of view. For example, a 90mm 'portrait' lens with an aperture of f/2.5 might give a beautiful portrait on 35mm film but the apparently 'equivalent' 60mm lens on an APS camera would not give anything like the same result, at the same aperture setting. This is because the depth of field depends on the physical size of the lens aperture and not its size relative to the focal length, which is what an 'f/' number means. Whereas f/2.5 on a 90mm lens represents a physical aperture of 36mm diameter, the physical aperture at 60mm focal length is only 24mm diameter, at the same 'f/' number. Hence, the results are different and this effect becomes more extreme as the sensor size of the digital camera is reduced.

When the 'depth of field' factor is taken into account, the potential size advantage of using a smaller sensor tends to disappear, for the portrait photographer, because the need to maintain the physical size of the aperture means that the overall size of the lens cannot be reduced, even if its focal length is shortened. This factor has led to some extreme lens designs, such as the large f/0.95 lenses for the 'micro four thirds' system. In this context, a Voigtlander 42.5mm f/0.95 lens on a 'micro4/3' camera is needed to achieve 'equivalent' depth of field to an 85mm f/1.8 lens on 35mm film.

The opposite can also be true, in applications where a large depth of field is required, such as macro photography One of my interests is photographing miniature scenes associated with my model railway. For this application, a 'compact' camera with a very small sensor provides an overall sharpness that is difficult to achieve with larger cameras.

The following pair of photos was taken from the same viewpoint, with lenses of 'equivalent' focal length, in terms of angle of view, both set to an aperture of f/5.6. In the case of the Lumix FZ200 (small sensor) camera, the sharpness of the image extends to objects in the background whereas the Olympus E-M5 only shows the engine in the foreground in sharp focus.


Aperture comparison

Comparison of depth of field for different sensor sizes at the same f/ number
left: Olympus E-M5 (focal length 45mm) right: Lumix FZ200 (focal length 16.4mm)
Both photos: f/5.6, ISO 200


ISO 'Speed'

In the case of film, the sensitivity or 'speed' of a film is determined by the manufacturing process. When digital sensors were introduced, which produce an electrical signal proportional to the amount of light falling on the sensor, it was realised that the effective sensitivity could be altered, simply by varying the amplification applied to the electrical signal. This is analogous to the volume control on an audio system and, in the same way that turning up the volume too much increases the background 'hiss' or 'noise', so excessive amplification degrades the final image quality. It follows that, in spite of the provision of an 'ISO' control on most digital cameras, there is an optimum setting, which makes best use of the overall 'dynamic range' of the sensor, which is the range from the amount of light that just provides a measurable signal to the amount of light that saturates the sensor.

There have been many manufacturing improvements that have increased the sensitivity of digital sensors to low light levels but the upper limit is determined by the capacity of each cell, or pixel, on the sensor, to hold the electrons generated by the light falling on the cell. This capacity is determined by the surface area of each pixel and, since the total size of the sensor is fixed, this capacity depends on overall sensor area. Hence, a smaller sensor has an intrinsically lower sensitivity to light than a larger one and the output will require more amplification, to achieve an 'equivalent' ISO value. In a sense, choosing a smaller sensor is akin to choosing a finer-grained film, which inevitably offers a lower ISO rating. This is another factor to be considered, when deciding if two cameras, with differently-sized sensors, are 'equivalent'.

In the case of my own cameras, the Olympus E-M1 sensor has an area of 225mm2, against only 28mm2 for the Lumix FZ200. The light-gathering capability of the Olympus can therefore be expected to be around eight times greater than that of the Lumix. This is equivalent to about 3 stops (3 EV) in terms of relative exposure. Thus, although the lens of the Lumix, with a maximum aperture of f/2.8 seems reasonably 'fast', the Olympus sensor should be able to achieve similar image quality with a lens of only f/8 aperture.  This leaves the opportunity for much better low-light performance, if an f/2.8 lens were used on the Olympus. In this sense, the f/2.8 lens of the Lumix could be said to be 'equivalent' to only an f/8 lens on the Olympus. Alternatively, the ISO setting of the Lumix must be kept down to 100, to match the image quality of the Olympus at ISO800.

The following pair of cropped photos are of an engraving of a flower petal, using the same exposure settings on two different cameras.  The reduction in image quality from the smaller sensor of the Lumix FZ200 can be seen in the right-hand image, where much of the fine detail in the engraving has been lost in the 'noise'. (both images processed from RAW files to eliminate in-camera JPEG processing)

grain comparison

Comparison of image quality for different sensor sizes at the same ISO setting
Left: Olympus E-M1 Right: Lumix FZ200
Both photos: 1/200s @ f/2.8 ISO 800

Conclusions

As usual, choice of a camera involves compromises. The use of small sensors in early digital cameras was largely dictated by manufacturing costs. As the technology matured, costs fell and photo sensors began to be included in most smart phones. The 'compact camera' market responded by favouring cameras with extended-range zoom lenses, which could select a small angle of view, for a 'telephoto effect', without the size and weight associated with previous designs.

Cameras with 'full frame' sensors tended to be large and heavy DSLRs, which could continue to use the lenses produced for film cameras and were widely used by professional photographers. As the manufacturing costs have fallen, however, larger digital sensors have found their way into more compact cameras, which are more like the old 'compact' film cameras. For applications where long lenses are not required, such as street photography, these large-sensor compact cameras have advantages, both in terms of light-gathering capability and overall image quality. Large sensors also offer advantages to the portrait photographer, because they make it easier to separate the subject from the background.

In the 'middle ground', the 'micro 4/3' system currently offers a good range of lenses, while being lighter than most DSLRs. This system suits me very well, providing a useful compromise between system weight and general image quality, especially in the context of natural history photography, where telephoto lenses are important.

When I want to photograph my railway models, however, I turn to my Lumix FZ200, because I can achieve a good depth of field at an aperture setting compatible with reasonable lighting levels, either flash or daylight.

For other applications, careful consideration of all the factors I have described in this article may lead to different decisions.  I should also point out that other factors influence the choice of a particular type of camera, including speed of operation, layout of the controls, or battery life, as a few examples.


Mike Flemming
October 2016

Return to Home Page